Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
0% found this document useful (0 votes)
123 views

Linprog 1 Sol

The document describes two linear optimization problems. Case 1 aims to maximize profits from producing two radar detector models given constraints on available components. The optimal solution is to produce 333.33 units of the SpeedBuster model for $45,333.33 in profits. Case 2 involves maximizing profits from three paper grades given a €100,000 budget constraint. The optimal solution is to produce 8,333.33 tons of paper grade Y for $333,333.30 in profits.

Uploaded by

yoursones
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
123 views

Linprog 1 Sol

The document describes two linear optimization problems. Case 1 aims to maximize profits from producing two radar detector models given constraints on available components. The optimal solution is to produce 333.33 units of the SpeedBuster model for $45,333.33 in profits. Case 2 involves maximizing profits from three paper grades given a €100,000 budget constraint. The optimal solution is to produce 8,333.33 tons of paper grade Y for $333,333.30 in profits.

Uploaded by

yoursones
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Linear Optimization 1

Case 1
Dear students, please read James Evan’s book Business Analytics, 3rd ed., Chapter 13 on Linear
Optimization.
Solve the following problem using QM.
Valencia Products makes automobile radar detectors and assembles two models: LaserStop and
SpeedBuster. The firm can sell all it produces. Both models use the same electronic components. Two
of these can only be obtained from a single supplier. For the next month, the supply of these is limited
to 4,000 of component A and 3,500 of component B. The number of each component required for
each product and the profit per unit are given in the table. The company wants to determine how many
of each model should be produced for the next month to maximize profit.
1. Identify the decision variables, objective function, and constraints in simple verbal
statements.
2. Mathematically formulate a linear optimization model.
3. Solve the problem using QM, and interpret your results.

Jawaban case 1
Decision variables:
• the number of LaserStop detectors to produce for the next month
• the number of SpeeBuster detectors to produce for the next month
Objective function:
To maximize profit by selling LaserStop and SpeedBuster.
Constraints:
• The number of components A used cannot be more than 4000
• The number of components B used cannot be more than 3500

Mathematical model:
Decision variables:
X1 = the number of LaserStop detectors to produce
X2 = the number of SpeedBuster detectors to produce

Objective function:
Max Z = 124X1 + 136X2
Constraint:
• Limit of component A:
18X1 + 12X2 <= 4000
• Limit of component B:
6X1 + 8X2 <= 3500
• Nonnegative constraints:
X1 >= 0, X2 >= 0

Linear programming model, as used as QM inputs

QM outputs

Interpretation: Optimal solution will be achieved by producing 333.33 SpeedBuster, which will give
the total profit of $45333.33.

Tambahan penjelasan
Pembahasan berikut tidak wajib disertakan dalam jawaban mahasiswa, melainkan hanya untuk
membantu mahasiswa memahami interpretasi dari output masalah linear programming.
Interpretasi output graph
Area berwarna biru merupakan feasible region. Artinya, semua titik (X1,X2) yang jatuh dalam area
ini memenuhi semua constraints, namun kita harus mencari titik yang menghasilkan solusi paling
optimal. Dalam kasus ini, solusi optimal adalah di titik (0,333.33). Titik optimal selalu berada di
sudut-sudut feasible region.
Garis merah menggambarkan isoprofit line. Artinya, semua nilai (X1,X2) di sepanjang garis ini akan
menghasilkan profit yang sama, namun kita harus mencari titik yang feasible.

Binding constraint
Binding constraint adalah constraint yang nilai slack/surplusnya = 0. Slack/surplus adalah nilai pada
persamaan constraint untuk membuatnya sama dengan nilai kanan (RHS, atau Right Hand Side).
• Slack jika constraint menggunakan tanda <=
• Surplus jika constraint menggunakan tanda >=
Nilai slack/surplus dapat dilihat pada windows Ranging dari output QM.

Component A merupakan binding constraint, karena slack = 0 (persamaan kiri sama dengan kanan).
Slack = 0, artinya semua sumberdaya sudah terpakai.
Surplus = 0, artinya tepat memenuhi syarat minimum.
A binding constraint is one where some optimal solution is on the line for the constraint. Changing
any of these constraints would either make the solution infeasible or not optimal.
Pada problem maximisasi dengan binding constraint <=, jika constraint kita naikkan, maka solusi
menjadi tidak optimal karena masih ada slack, sehingga nilai objective function masih bisa dinaikkan.
Jika constraint diturunkan, maka solusi akan melebihi constraint sehingga menjadi tidak feasible.
Interpretasi nilai dual
Component A memiliki nilai dual 11.33. Ini artinya, jika component A bisa ditambah 1 unit, akan
menambah total profit sebesar $11.33 ke nilai objective function. Perhatikan bahwa component A
adalah binding constraint, jadi sudah terpakai semua.
Untuk component B, ada slack, belum terpakai semua, jadi jika komponen B ditambah 1 unit pun,
tidak akan menambah profit, karena yang sekarang pun tidak habis terpakai.

Dual problem membalik matriks primal problem (transpos matriks).


Pada primal problem:
• Decision variables adalah kolom matrix
• Constraints adalah baris matrix
Primal problem ke dual problem:
• decision variables primal menjadi constraints dual (decision variables menjadi baris matrix)
• constraints dual menjadi decision variables primal (constraints menjadi kolom matrix)
• objective maximisasi primal menjadi minimisasi dual, dan sebaliknya
• koefisien objective function primal menjadi RHS constraint dual
• RHS constraint primal menjadi koefisien objective function dual
• Tanda <= pada kendala primal menjadi >= pada kendala dual, dan sebaliknya
Window dual pada output QM memperlihatkan bentuk primal problem dan dual problemnya.
Interpretasi ekonomis dari dual problem:
$4001 dan $3500 mencerminkan harga maksimum yang bersedia kita keluarkan untuk mengkonsumsi
setiap unit sumberdaya komponen A dan komponen B. Maka interpretasi dari objective function
adalah, meminimalkan biaya untuk mengkonsumsi sumberdaya, ekivalen dengan memaksimalkan
profit.
Interpretasi constraints:
• Constraint LS: minimum profit yang diharapkan dari produk LaserStop adalah $124
• Constraint SB: minimum profit yang diharapkan dari produk SpeedBuster adalah $136

Case 2
Dear students, please read James Evan’s book Business Analytics, 3rd ed., Chapter 13 on Linear
Optimization.
Solve the following problem using QM.
A paper mill produces three grades of paper: X, Y, and Z. The mill has a budget of €100,000 to invest
in the production of these three different types of paper. The cost per ton and expected profit over the
next two years is given in the table. The company wants to determine how many of each paper grade
should be produced to maximize net profit.
1. Identify the decision variables, objective function, and constraints in simple verbal
statements.
2. Mathematically formulate a linear optimization model.
3. Solve the problem using QM, and interpret your results.

Jawaban case 2
Decision variables:
Number of paper grade X, Y, Z to produce (in ton) for the next two years
Objective function:
Maximize total profit from all three types of papers
Constraints:
total cost to produce all paper cannot exceed €100,000

Mathematical model
Decision variables:
X = number of paper grade X
Y = number of paper grade Y
Z = number of paper grade Z
Objective function:
Maximize Profit = 30X + 40Y + 20Z
Constraints:
Budget:
10X + 12Y + 7Z <= 100,000
Nonnegative:
X >= 0, Y >= 0, Z >= 0
Input QM

Output QM

Interpretasi:
Solusi optimal dicapai dengan membuat paper Y sebanyak 8333.33 unit, yang akan menghasilkan
profit sebesar $333,333.3

Tambahan penjelasan
Budget adalah binding constraint, artinya budget sudah terpakai semua, tidak ada slack. Andai budget
bisa dinaikkan 1 unit, dapat menambah profit sebesar $3.33 (terlihat dari nilai dual).
Tidak ada graph karena ada 3 decision variables

You might also like